Research Advances 1. STIM2 mediates STIM1 conformational switch to promote STIM1/Orai1 activation at minimal ER-Ca2+ depletion: There is considerable evidence that, at low ER-Ca2+, Ca2+ dissociates from the N-terminus of STIM1, which initiates STIM1-STIM1 association and disruption of the intramolecular C-terminal CC1-CC3 interaction, and promotes the activated conformation required for gating Orai. These conformational changes in STIM1 are also critical for its clustering in ER-PM junctions with Orai1. We have now examined how STIM2 facilitates STIM1 clustering and gating of Orai1 at relatively high ER-Ca2+, i.e. when the STIM1-EF hand domain is expected to be loaded with Ca2+. We find that STIM2 expression causes recruitment and activation of endogenous Orai1 in resting cells. Co-expression of STIM1 with STIM2 (wild type; or gating- deficient mutant, STIM2-L485H) promoted assembly of STIM1/STIM2/Orai1 and increased Orai1 function (Figure 1B). In contrast, STIM1-F394H (lacking the ability to gate Orai1) or STIM1-R426L (with stabilized CC1-CC3 interactions) did not increase Orai1 function when expressed with STIM2. Importantly, STIM1-ES122 (with stabilized N-terminal EF-SAM domain) induced similar Orai1 activation as did STIM1. Thus, STIM2 interaction with STIM1 causes changes in the STIM1-CC1-CC3 domains that lead to Orai1 activation. We confirmed this by using the STIM1 conformational sensor (C-OASF-Y)-FRET. STIM2, but not STIM2-SOAR, decreased the FRET of C-OASF-Y. Together these findings reveal that under conditions in which ER-Ca2+ depletion is insufficient to trigger a response in STIM1, a critical STIM2-STIM1 interaction controls STIM1 conformation as well as its assembly with, and activation of, Orai1. Our results show that at relatively high ER-Ca2+, STIM2-STIM1 interaction promotes the active STIM1 conformation required for Orai1 activation. 2. PIP2-dependent cytoskeletal remodeling is involved in assembly and function of Orai1-STIM1 complexes: Reports suggest that PIP2 is essential for Orai1-STIM1 assembly in ER-PM junctions and coordination of cytoskeletal remodeling that might impact SOCE. Further, septin has been linked with organization of PIP2 in these junctions. Our studies reveal that a PIP2-dependent actin remodeling complex containing CDC42/N-WASP/WAVE/ARP2/3 is recruited to ER-PM junctions during SOCE, and affects the assembly and function of STIM1-Orai1. Depletion of PIP2 or knockdown of septin impaired STIM1-Orai1 assembly, SOCE and NFAT activation, together with a decrease in recruitment of CDC42. Knockdown of CDC42 and its interacting proteins, N-WASP/WAVE and ARP2 reduced STIM1-Orai1 clustering and function, but not Orai1 activation by STIM1 C-terminus or STIM1K. Importantly, TIRFM demonstrates a distinct reorganization of actin around STIM1/Orai1 clusters, which was dependent on CDC42 and ARP2/3. These findings suggest that PIP2 mediates recruitment of an actin remodeling complex within ER-PM junctions, which promotes/stabilizes a functional STIM1-Orai1 complex. The effects of cytoskeletal remodeling on STIM2/Orai1 assembly/regulation as well as TRPC1 trafficking need to be determined. 3. STIM2 promotes coupling of SOCE to the activation of NFAT: There is much interest in mechanisms that regulate the assembly and stability of SOCE components in ER-PM junctions. Our current studies examine the role of STIM2 in the assembly of Orai1/STIM1 in ER-PM junctions. By using an ER-marker protein to denote ER-PM junctions, we find that Orai1 does not induce ER recruitment into ER-PM junctions on its own, while clustering of STIM1, STIM2 or STIM1+STIM2 following store-depletion cause recruitment of ER into the clusters. In the case of STIM2, this is achieved at low agonist. Plasma membrane PIP2 have been shown to be importance for the recruitment of STIMs to ER-PM junctions and for SOCE. Phosphatase-induced depletion of PIP2 levels in the plasma membrane significantly decreased SOCE and clustering of both STIM1 and STIM2. Knocking down endogenous STIM2 also significantly reduced SOCE and NFAT activation, whereas loss of STIM1 abolished both. While co-expression of both Orai1 and STIM1 was sufficient to generate ICRAC, STIM2 is required for slow Ca2+-dependent inactivation of ICRAC. Similar to full-length STIM1, co-expression of Orai1 with STIM1K (mutant with polybasic domain deleted) also generated ICRAC and SOCE. However, the Orai1+ STIM1K-mediated Ca2+ entry did not activate NFAT. When STIM2 was co-expressed, it colocalized with Orai1 and STIM1K clusters, and also rescued NFAT activation. In summary, our data suggest that STIM2 plays an important role in modulating Orai1 channel function and coupling Orai1-mediated Ca2+ entry to the NFAT signaling pathway. Ong et al., manuscript under preparation. 4. Radiation triggers a TRPM2-dependent mitochondrial pathway to promote STIM1 cleavage by Caspase 3 and loss of SOCE. We have now addressed two major questions: (a) What is the underlying defect in acinar cells that causes loss of salivary fluid secretion? and (b) Is TRPM2 function associated with this defect? Our findings elucidate for the first time that SOCE, which is critical for salivary gland fluid secretion, is compromised by IR. Further, acinar cells from TRPM2+/+, but not TRPM2-/-, mice display an increase in Ca2+mt and ROSmt, a decrease in mitochondrial membrane potential and activation of Caspase 3, that is associated with persistent reduction in STIM1, SOCE and loss of salivary gland fluid secretion. In a salivary gland cell line, silencing Mitochondrial Calcium Uniporter (MCU) or treatment with TRPM2 or Caspase 3 inhibitors prevented IR-induced loss of STIM1 and SOCE. Importantly, adenovirus-mediated expression of STIM1 in vivo in salivary glands of IR-mice increased SOCE and fluid secretion. Thus, IR triggers a TRPM2-dependent mitochondrial pathway to promote STIM1 cleavage by Caspase 3 and loss of SOCE. These data provide the first mechanism that can fully account for IR-induced loss of salivary gland function. (Liu et al. 2017, Sci Signal. 2017 Jun 6;10,482). 5. Adenovirus mediated expression of hAQP1 in irradiated mouse salivary glands increases fluid secretion by enhancing carbachol-induced cell volume decrease. Agonist-stimulated cell volume decrease is a read-out for the secretory activity of acinar cells and correlates well with saliva secretion in animal models and human subjects. Compared to the sustained decrease in volume of acinar cells from non-IR mice, CCh induced a transient and attenuated decrease in cell volume of salivary gland acinar cells from IR-mice (2 and 8 months post-IR). The attenuation of CCh-induced cell volume change can be accounted for by the defect in CCh-induced Ca2+i increase (described above). Despite this defect, expression of hAQP1 in glands of mice post-IR, led to recovery of CCh-induced volume changes, with increase in the initial rates of cell volume reduction, but not the amplitude. Together the data suggest that expression of hAQP1 increases the water permeability of acinar cells, which underlies the recovery of fluid secretion in salivary glands functionally compromised by IR. Although Ca2+ signaling defects in acinar cells might preclude full recovery of saliva secretion, this is still a useful strategy for treatment. These data establish the physiological basis for the AdCMV-AQP1 gene therapy strategy. (Teos et al. Gene Ther. 2016 Jul;23(7):572-9).